Cytomegalovirus (CMV), which infects half of the US population, establishes a persistent infection for the life of the person and is of particular riskto solid organ and hematopoietic stem cell transplant patients and individuals with a compromised immune system. Natural killer (NK) cells, together with T cells, are required to control this persistent infection for the lifetime of the individual. We recently made the unexpected observation that human NK cells bearing a unique invariant activating CD94-NKG2C receptor can preferentially and specifically respond to CMV infection in the outbred human population. We had previously demonstrated that the CD94-NKG2C receptor recognizes HLA-E, an essentially non-polymorphic major histocompatibility complex protein that is expressed on all cells and tissues in the body. The expansion of CD94-NKG2C+ NK cells following CMV infection likely reflects the recognition of a peptide from CMV or the infected host cell expressed in the context of HLA-E. In this project, we propose to use a novel and elegant technique to determine the changes in the peptide repertoire of HLA-E in CMV-infected cells by creating soluble, secreted HLA-E molecules that can be easily purified for elution of bound peptides followed by comparative mass spectrometry. Aims of this project are: 1) To determine the nature of the peptides bound to HLA-E in uninfected versus CMV-infected cells. We will test the hypothesis that CMV-infected cells will demonstrate an altered HLA-E peptide repertoire, including the generation of HLA-E proteins displaying CMV-encoded peptides, as well as CMV-induced host-encoded peptides. 2) To measure the binding of CMV-induced HLA-E -peptide complexes for the activating CD94-NKG2C versus inhibitory CD94-NKG2A receptors and test the ability of these HLA-E-CMV-induced peptide complexes to stimulate primary human NK cells expressing the CD94-NKG2C receptor. Identifying high affinity or preferential HLA-E ligands for the CD94-NKG2C receptor will provide the foundation for future studies to characterize the biological consequences of recognition of these novel ligands. Vaccines for CMV are an unmet medical need and the identification of these novel CMV-induced HLA-E-peptide antigens may provide new therapeutic strategies for the treatment or prevention of CMV infection.